OpenDR successfully organized on October 1st 2021, under the leadership of PAL Robotics, an IROS 2021 workshop on “Open Deep Learning Toolkit for Robotics: Towards Democratizing Artificial Intelligence”. The workshop, which was held virtually and was well attended, was structured in two sessions, each followed by a lively panel discussion involving the session speakers and a moderator. Members of the OpenDR consortium contributed to the workshop with interesting talks on a variety of topics:

  • Deep Learning and AI Applications in the Webots Robot Simulator (Stefania Pedrazzi)
  • Robotic grasping in Agile Production (Roel S. Pieters)
  • Deep Learning models for efficient video-based action recognition (Alexandros Iosifidis)
  • Learning Efficient Representations for Perception, Tracking, and Localization (Abhinav Valada)
  • How to benefit from learning in predictive control (Laura Ferranti) 
  • Business perspectives and applications of OpenDR for AGROINTELLI (Ole Green)

Moreover, the workshop included talks from prominent roboticists namely:

  • Using Open Deep Learning Toolkits in ROS/ROS2 (Francisco Martin Rico, Rey Juan Carlos University Spain)
  • European Training Network on Personalized Robotics as Service Oriented applications (Silvia Rossi, University of Naples “Federico II”, Italy)

Apart from their live talks, speakers made available longer,  recorded versions of their talks as well as the talk slides. These videos will gradually become available in the OpenDR YouTube Channel. Francesco Ferro and Sofia Battilana (PAL) played a key role in the workshop, by coordinating the flow and providing opening and closing remarks.